Do road investments lead to economic growth?

Road improvements lead to benefits in the form of e.g. reduced travelling time, improved traffic safety and reduced emissions. These improvements do not only benefit the actual road users, but they are also "spreading" into the local community or neighbouring communities through several types of effects e.g. enlarged labour markets. There may be "wider" benefits, meaning that the total net benefits of the project are greater than the sum of net benefits of the road users. Actually the question is not whether these "wider" benefits exist, but whether they are of any practical importance or if they might as well be ignored in ordinary economic evaluation. During several decades economists have tried to investigate the hypothesis with varying results depending on model as well as on data. The present paper follows up our two earlier studies, where we have tried to establish whether road investments contribute to economic growth, which our earlier studies give little support for. Previously we have analysed data for industry sector or for geographical regions. This time we have sufficient data for analysing industry sector within each region. Our approaches are inspired by an article by John G. Fernald presented in 1999. Four models are analysed, based on data from Statistics Norway, for the period 1997-2005. The models have the same design, only the size of time-lags varies. The analyses show that out of four models only one produces significant results. In this model productivity growth is lagged two years behind road investments to allow for the improvements to be completed. What we find is that significance is high, but the size of the productivity coefficients is rather small, just around 1 2 percents, depending upon the cost shares. Models with other time-lags produce insignificant and diverging results. The time variable is highly significant with a positive sign, thus indicating that road investments are becoming more productivity enhancing over time.